1 paper
Vanessa Liao, Syed Shariyar Murtaza, Yifan Nie +1
A common way to use large pre-trained language models for downstream tasks is to fine tune them using additional layers. This may not work well if downstream domain is a specialize…